The Problem with a Variable Tax on Gasoline
“Let’s make our demand for gasoline even more inelastic.”
Other economists suggest a gasoline tax that would increase as gasoline prices fall toward a predetermined floor and decrease as pump prices rise above it. This way, OPEC countries would be unable to manipulate prices lower so as to destroy the returns of persons risking capital in alternative-energy investments, which the cartel has adeptly done in the past.
This is from Jim McTague, “Obama is Humbled by the Market,” Barron’s, March 24, 2012. Greg Mankiw linked to it this morning. I’m guessing that what caught Greg’s attention was the article’s advocacy of a tax on gasoline. I won’t bother revisiting that well-worn debate.
Instead, I want to focus on a paragraph that I bet Greg didn’t focus on, the one I reproduced above. I wrote about a similar proposal, a variable import fee on oil, in two articles (here and here) in the Energy Journal in 1989. A variable gasoline tax is not quite the same, but it would have a similar unintended, but completely predictable, consequence.
The argument I’m about to give rests on the assumption that the world oil market has an entity in it, OPEC, that has some market power. If you don’t believe that, then you won’t find the rest of the argument convincing. But notice that McTague does believe it. Check the quote from his article above.
What constrains OPEC from raising price above whatever level it’s at? Three things: (1) production by non-OPEC suppliers, (2) strains within OPEC as members find it increasingly attractive to “cheat” on the cartel agreement by cutting price a little, and (3) the size of the world’s elasticity of demand.
It’s on (3) that I want to focus. The United States consumes about 25% of world oil and (I don’t have it handy) more than 25% of world gasoline. There’s some elasticity of demand without the proposed variable tax.
But now assume that the government says that the target price of gasoline, in 2012 dollars, is $4.00 a gallon. For simplicity of exposition, assume a combined federal and state gasoline tax of 50 cents per gallon at that price. So the price minus tax is $3.50 a gallon. What happens if OPEC cuts the price of oil, which would cut the price of gasoline? Normally what would happen is that OPEC would sell more oil used to make U.S. gasoline. But if OPEC cuts the price of oil, the price of gasoline as seen by the U.S. consumer would not be allowed to fall. If the price minus tax would have been, say, $3.25, then the tax would rise to 75 cents. The consumer would see the same price of gasoline he would have seen and, therefore, will not change his quantity demanded. So that one source of elasticity of demand that previously confronted OPEC–the U.S. gasoline consumers’ elasticity of demand–no longer exists. The government has created a perfectly inelastic demand curve for gasoline. That, on the margin, reduces OPEC’s incentive to cut price.
Similarly, if OPEC is contemplating a price increase that would normally drive the price of U.S. gasoline above $4.00 a gallon, one thing constraining it is the fact that U.S. consumers would cut back the amount of gasoline they demand. So, for example, if OPEC raised the price of oil so that the price of U.S. gasoline, including the normal 50 cent tax, would be $4.25, then OPEC would know that U.S. consumers would cut back somewhat. But now OPEC would know that the price the U.S. consumer would see is only $4.00 a gallon because at the higher price of crude that would normally yield a market price of $4.25, the U.S. government would cut the gasoline tax to 25 cents a gallon to keep the price of gasoline at $4.00 a gallon. So why not raise the price of oil?
Of course, there are a few “why nots:” factors (1) and (2) mentioned earlier, and the elasticity of demand of the rest of the world and of U.S. consumers of non-gasoline refined products. But the variable gas tax eliminates the U.S. demand for gasoline as one of these factors.
So, ironically, but completely predictably, a policy allegedly aimed at undercutting OPEC’s monopoly behavior would actually strengthen it.